A Hybrid Scheme for Feature Selection of High Dimensional Educational Data

Now a days, in educational data sets, the gigantic amounts of records can be the problem in generating good quality data. Lately, a lot of educational researchers are using the data mining methodology to analyze the data. But, several research studies focus on the selection of right learning algorithm rather than carrying out the feature selection on data. Therefore, the dataset has problem like high computational complexity and performing classification on such data requires a lot of computational time. This paper will give a summary of the feature selection methods which are being used for analysis of features of data. The proposed hybrid methodology is a combination of both feature selection and wrapper based technique, which helps to enhance the quality of students' data set.